GAN-Based Data Imbalance Techniques for ECG Synthesis to Enhance Classification Using Deep Learning Techniques and Evaluation

The prompt detection of irregularities on Electrocardiograms can be crucial in patient monitoring, deep learning-based strategy that combines segmentation and classification as one procedure has recently been effectively applied to ECG signal analysis. In the gathered ECG signals, it frequently results in an unbalanced ratio of small and large cases of numerous abnormalities. The deep network is unable to learn how to recognize minority groups because of this data imbalance. Various physiological signal analyses, medical imaging datasets have all been balanced using GANs. GAN and Deep Learning integrate well together in the healthcare, particularly for the recognition and characterization of ECG patterns. In order to improve categorization and efficiency, this study presents a comprehensive overview of the GAN strategies for data imbalance and deep learning strategies for physiological signal classification. With Cardio GAN, CNN obtained accuracy greater than 99.0 percent and performs superior classification.